1.7.2 Is the world objective or subjective?
Philosophers have speculated for millennia on the following question. Is the world an objective entity independent of our subjective knowledge of it?
\We will not explore the very long history of philosophical debate but focus on some key issues pertinent to researching the social world.
First, there is no definitive resolution to the problem of whether the world exists 'out there' independent of our minds?. The famous example of Dr Johnson trying to prove to Bishop Berkeley the existence of an objective external world, by kicking a stone and saying he can feel the pain, is inconclusive. The act of kicking, the pain felt and even the conversation could all be in Dr Johnson's mind.
Second, having said that, most social scientists adopt a view that there is, in effect, an objective social world independent of themselves of which they are a part. However, as we noted above (section 1.6.1), the nature of the social world, its ontological make up, is viewed in various ways, which impact on how we research it.
Third, although social scientists tend to assume an objective social world, there is far less implicit agreement as to whether it can be known objectively.
1.7.3 Is the world objectively knowable?
What do we mean by objective knowledge? Simply put, objective knowledge exists independent of the knower, that is, it is knowledge that is 'out there': that anyone can 'pick up'. The Earth orbits the sun, is an example, for some, of objective knowledge (although 500 years ago the Earth was the centre of the universe!).
Subjective knowledge is rather more complicated. It can mean knowledge known by an individual (or group) but generally unknown and therefore not part of the knowledge base of a subject discipline. Often this might be construed as a person's belief or opinion.
More important, from a social research perspective, subjective knowledge is knowledge as interpreted or understood by a researcher but which cannot be presented as definitive knowledge independent of the subject who knows it.
For example, a researcher may know, on the basis of an interpretation of a conversation with a subject, what meaning a social encounter has for the subject. This is not something anyone else can know directly as the conversation was personal and the interpretation that of the researcher. This cannot be exactly replicated either. Another researcher could have a similar conversation with a similar subject and make a similar interpretation. However, these two events can never be identical as the historical and social context will have changed and the nature of the dialogue will, at the very least, have subtle differences.
There are significant differences in opinion, among social scientists, about the extent to which the social world can be objectively known. Indeed, the discussion also applies to knowledge of the natural world, although many scientists assume that knowledge of the natural world exists to be discovered, the history of the philosophy of science shows that this is a heavily contested contention.
The following discussion is conceptually complex and for students new to social research or social philosophy, it may be easier to assimilate the debate once you have read Part 2 or undertaken some social research.
220.127.116.11 Discovering independent objective knowledge
There are those who take the view, or act as though they ascribe to the view, that the world exists independently of ourselves and that knowledge of it is also independent and waiting to be discovered.
There are few social researchers nowadays who would admit to this position philosophically but many who implicitly adopt the approach in practice. One example of a clear positivist stance on objective knowing can be found in the discipline of international relations. Hans Morgenthau (1967, p.4) described as the founding father of realist international relations (Guillaume, 2002), stated his belief that:
politics, like society in general, is governed by objective laws that have their roots in human nature. In order to improve society it is first necessary to understand the laws by which society lives. [Realism] must also believe in the possibility of developing a rational theory that reflects, however imperfectly and one-sidedly, these objective laws. It believes also, then, in the possibility of distinguishing in politics between truth and opinion—between what is true objectively and rationally, supported by evidence and illuminated by reason, and what is only a subjective judgment, divorced from the facts as they are and informed by prejudice and wishful thinking.
For Morgenthau, observation, experimentation, falsificationist testing and the judgement of history are some ways to test that theories correspond to the truth of external reality. The approach has been 'adopted by most international relations scholars since then, whether "realist" or not. This monistic metaphysic implies that as there is one world "out there", there can only be one true (scientific) perspective corresponding to this sole reality' (Guillaume, 2002).
The philosophical debate underlying this positivist assertion of the correspondence of knowledge to objective reality is rather more sceptical. John Locke was well aware that two people could have different subjective experiences of the same phenomenon. Locke demonstrated that the same person could experience the same thing in two different ways: placing one hand in a bucket of cold water, the other in hot water and then transferring them both to a bucket of tepid water and the water feels simultaneously hot and cold. Locke, thus, suggested that objective reality had primary (or intrinsic) properties that we could apprehend through science, even though our perception of these properties were secondary and subjective.
Emanual Kant, agreed that an object had specific properties, which he called the-thing-in-itself (Ding an sich, in German) but we cannot know its true nature other than that it has one. Scientific knowledge for Kant is systematic knowledge of the nature of things as they appear to us subjects rather than as they are in themselves. Thus, the Ding an sich can never been known objectively, in the purist sense of the word. Therefore, the only plausible sense for the term 'objective' would be to apply it to judgments about which there is agreement; preferably universal agreement. We will come back to intersubjective agreement below.
18.104.22.168 Constructing knowledge to fit frameworks
There are others who argue that the world can only be known through the way we construct knowledge about the world. That is, knowledge is constructed through human agency. Individuals construct knowledge of the world by fitting new experiences into frameworks that they use to make sense of the world: knowledge of the social world is individualised and transitory.
A rather less radical version of this approach suggests that the construction is not simply individualistic but that the devices we use to construct and communicate knowledge actually create our image of the world. For example, the language we use to communicate our notion of the world affects the concepts we have of the world and thus our knowledge of it. (We will explore such issues in subsequent Sections).
22.214.171.124 Socially-constructed knowledge
There are others who argue that knowledge of the human world is created by the way we develop our knowledge of it but that the way we construct knowledge and the nature of the world are interdependent. The world we live in affects the way we understand the world: it provides us with the concepts, beliefs, ideologies and other frameworks for making sense of the world, but the tangible conditions we live in, the things we do also affect the way we construct knowledge. Furthermore, that knowledge is not individual but social, we do not live in isolation, the frameworks we use and the conditions we live in are socially constructed and so is the knowledge we have of the world.
126.96.36.199 Objectivity and epistemology
In the first view, (that there is independent knowledge, 188.8.131.52, above) knowledge is assumed to be objective and discoverable.
In the second view (that knowledge fits frameworks, 184.108.40.206, above), knowledge is subjectively constructed but communicated and 'objectified' by being shared, although it is transitory.
In the third view, (socially-constructed knowledge, 220.127.116.11) knowledge is social and objective although contingent upon social (physical and ideological) conditions.
These relate to the epistemological outline above (Section 1.6.2). Positivism and the search for causes is compatible with an objectivist view of discoverable knowledge, although most positivists adopt a less fundamentalist view about external objective knowledge.
Phenomenology and the search for meanings, in its many versions, is compatible with the individual construction of knowledge, although few phenomenologically-inspired social researchers would take an unadulterated radical view of the subjective nature of knowledge.
Critical social research, in its search for historically and structurally-located knowledge, is compatible with the social constructive approach to objective knowledge.
These three perspectives are explored in more detail in Part 2 and linked to real examples of social research.
1.7.4 What are objective data?
The philosophical discussion outlined above and the predominant view in social science would have it that knowledge of the social world is mediated by the knowing subject—the researcher. At the very least, the argument goes, the social world is unlike the natural world because the subjects of the social world are reflective, conscious knowing people who can respond to the knowledge that is known about them, which clearly is not the case of the natural sciences. A rock for a geologist is a rock. What we know about it may change but the rock does not.
In that case there can, in purist terms, be no objective data about the social world. There is, as noted above (Section 1.4.2) a tendency to talk about facts as thought they are objective data. However, 'facts' are theory-laden, a datum only makes sense in its context. So we cannot simply assume that 'facts' are objective data.
What tends to happen in practice is that, instead of a simple dichotomy of objective and subjective, data is seen as more or less objective. In practice, objectivity relates to the use of evidence to establish the credibility of a statement. This means that, for example, presupposition, hearsay, desire and prejudice constitute a subjective realm of knowledge, while the use of any empirical data offers the basis for an objective view (or 'more objective' view). In practice, one could argue that empirical data (rather than being objective per se) can provide evidence in support of a view.
1.7.5 What are objective methods?
The majority of the debate about objectivity and subjectivity in the natural and social sciences focuses on research practices and makes claims about the extent to which research practices are objective or subjective. Usually, the 'ideal' objective research practice is the controlled experiment (see Section 9). The 'scientific' experiment allows the researcher to control the whole situation and to measure changes in one variable in response to changes in another. Thus, controlled experiment is apparently accurate and reliable, in as much as someone repeating it would come up with the same results. It is also often assumed to be valid but the validity is restricted to the set of preconceptions that guide the experimenter (see section 8).
Before combustion was identified as an oxidising process, many experimenters carefully measured the amount of phlogiston that different substances emitted or absorbed when they were burned. Although reliable and accurate, the validity of what they were doing was entirely spurious (because phlogiston did not exist, although the theory of the time said that it did). (See section 9 on reliability and validity)
Debates about the objectivity of social science method do not treat subjectivity and objectivity as binary opposites. They presume, in effect, that there are degrees of objectivity. The binary view is glossed over, not least because it would be hard to make a convincing case that even the controlled experiment is entirely objective. The debate, in social science, over the objectivity of method hinges on the extent to which the researcher intervenes in the data collection process. This may involve the extent to which the data collected is dependent on the interviewer's interpretation, or the questions the researcher asks or the range and variability of circumstances that the researcher is able to investigate.
The 'ideal' objective method would be one that:
- minimises the impact of the researcher on the research subjects;
- ensures that all those about whom inferences are to be made are included in the research, or, if that is not possible, that a representative (non-biased) sample is used;
- collects data in the same way from, or about, each research subject;
- does not lead subjects, deliberately or inadvertently, to provide data that confirm the researcher's presuppositions;
- enables data collection that might either confirm or deny a hypothesis, presupposition or theory (that is, that does not restrict the data collection only to data that might confirm the researcher's prejudices.)
This ideal approach is not only assumed to be 'more objective' but also enables results to be generalised (see section 10 ). However, this ideal approach to method places a premium on surface data collection rather than in-depth enquiry.
18.104.22.168 Value-free research
Another dimension of the 'objective' approach is that the research should be independent of the values of the researcher.
There is widespread agreement that the term 'objectivity' is synonymous with such things as neutrality, impartiality, and disinterestedness; the objective observer, for instance, is able to give a reliable account of events precisely because she or he has no interest in the outcome and is able to make statements and render judgments regardless of their consequences. Apparently, we have managed to agree about what an objective observer is, even though we usually disagree about whether this or that person has in fact served as an objective observer in any given case. (Bérubé, 2005)
The subjectivity-objectivity debate has been complicated by its development in the context of a separate but overlapping discussion of the relationship between fact and value. The fact-value dichotomy has served to cloud the issue of objectivity.
The need for 'value-free' social research was highlighted in the work of Max Weber. He distinguished between what he called 'value relevance', which was whether the research was worth doing, and 'value-freedom', the perspective the researcher is supposed to adopt when undertaking research: that is, not distorting the outcomes for non-scientific reasons, such as personal preference or predilection.
Two issues emerge for Weber. First, 'fact' and 'value' should be kept separate and only the former should admitted to scientific analysis. This, as we have seen is problematic because of the theory-laden nature of observation. Indeed, one argument is that any observation is imbued with the observer's values, to some extent people see what they want to see.
Second, Weber argued that moral judgements may be construed as facts and therefore admissible to social science. For Weber, moral judgements are societal choices and thus 'social facts' or, alternatively, they are self-evident and objectively graspable, for example, cruelty is self evidently wrong and it is possible to empirically identify cases of cruelty. Weber's position is tenuous, as Lessnoff (1974) has argued: how does one distinguish between accurate and defective moral sense? The US Republican government at the start of the 21st Century implied that it was morally acceptable to torture suspected terrorists: much the same logic applied by the inquisition to potential heretics in the Middle Ages.
Over the years, Weber's position was adopted by and criticised by a variety of researchers. One milestone critique was that proposed by Alvin Gouldner (1961). He criticised the dominant structural functionalist tendency in American sociology of the 1950s arguing that the Talcott Parsons and Robert Merton, for example, did not adopt a value-free approach despite their claims. Instead, they were representing the politically conservative approach as ideologically-neutral and apolitical. Gouldner regarded this as intellectually dishonest as they were reproducing the status quo as though it were unproblematic and natural (see Section 22.214.171.124 on ideology). Gouldner pointed to another approach, which he labelled the sociologist as partisan (Gouldner, 1968), that proposed that sociologists should be explicit about the nature of the world they are studying, not be afraid to challenge the status quo and injustice and select the phenomena worth studying. The latter is clearly driven by social values. Although Gouldner had concerns with Becker's tendency to side with the outsider, Gouldner argues that being explicit about values is not the same as being subjective. Value-explicit research can be undertaken in an 'objective' manner. Conversely, hiding one's values does not make the research process any more 'objective'.
The premises one starts from are not neutral. Becker (1967), for example criticised the predominant approach to juvenile delinquency by arguing that it started from the premise that there was something wrong with juveniles. This prejudiced the way juveniles were researched. It might, he argued be just as reasonable to state the 'problem' as what is wrong with the parents.
We need to ask ourselves whether it is really possible to rid ourselves of value judgements. That is we need to ask if value-free research is possible. Instead of assuming, as many scientists do, that research can be carried out in a value-free way, we could instead ask researchers to be as explicit as possible about the values that have influenced their research. If we come to the conclusion, as many social researchers now do, that making value judgements is part of our human condition, then is it not pointless to try to eliminate values from the research process' If we come to the conclusion that it is pointless to try to eliminate values from the research process what should we do? May (1997) suggests that we should be turning the question round and asking what type of values do we base our judgements on? In this way, we are able to understand the judgements that researchers make and we are able to understand our own reasons for accepting or rejecting them.
Researchers who acknowledge the role of values in research accept that different people in society have different interests. For example, we come from different cultural and historical backgrounds and our experiences are mediated by such things as our ethnicity, social class, gender, sexuality and ability or disability. We behave in ways that will be likely to further our interests. These interests affect the ways in which we define the problem or issue that we wish to research.
126.96.36.199 Objectivity and the politics of research
Politics, as practiced, is an area where objectivity is assumed to be lacking. Politicians seem to regard objectivity as irrelevant and decisions are made on party lines; what one party proposes the other opposes, irrespective of any objective consideration. Whether this renders politics as a discipline incapable of objective knowledge is not the point.
What it points to is that, in social and natural scientific research, decisions have a political element, that is, the decisions made, the approach taken, the resources allocated, the outcomes reported usually have an extra-scientific dimension. We discuss this in more detail in chapter X.
What this means, for the discussion of objectivity, is that it is naive to assume that scientific research of any kind can be accomplished in an objective vacuum. As Gouldner (1961) argued, politics is part of every aspect of our lives. Political neutrality is a myth.
Faced with the ubiquity of politics, one needs to adopt a critical reflective approach when examining any research. One might explore the influences on the research, what the terms of reference of the research were, who funded it, why certain methods were undertaken and what purpose was the research intended to fulfill.
Finally, as power is not evenly distributed throughout society we need to ask questions about who has the power to construct certain issues as a problem in the first place. In other words, the values themselves become the focus of the research. Feminist researchers, for example, have pointed to the ways in which sexist and racist values have influenced the research process. Thus, we need to explore why some areas are selected as being of interest whilst others are not. We also need to explore the uses to which research may be put; often this may be far from the researcher's original intentions.
The study of society involves researchers taking a self-critical approach to their values and exploring the impact that has on both the topics that are to be researched and the ways in which research is carried out. Even the dictum that science should be value-free is itself a value position.
1.7.6 Can theories be objective?
An objective theory is, at one level an oxymoron—that is, self-contradictory—because a theory is a speculation, relying on a subjective framing of a problem. There are, though, views that suggest theories are objective.
First, the positivist approach that a theory is objective if it is rational and logical (thus objectivity does not depend on correspondence to particular objects). Second, a theory is objective if one can (repeatedly) demonstrate the predicted outcome of theory (usually in a controlled setting, such as an experiment). Third, a pragmatic/ist approach that says the theory is objective if it is useful, i.e. it works in everyday situations.
Chalmers (1973), proposes a more fundamental objectivist approach. He does not see this as counter to his view of the theory-laden nature of observation (section 1.4.2). An objectivist approach (which he endorses) conceives scientific theories as autonomous, existing independently of individual or community opinion, notwithstanding the fact that the participation of scientists (as individuals or communities) is necessary for the development and generation of theories. This objectivist approach rejects the view that science is a set of beliefs, held either individually or communally. In this sense, science is a process without a subject. Chalmers view is that scientific theories are interrelated and bear a certain relationship to the available evidence, whether or not any scientist or community specifies it. Chalmers states that
Theories are coherent or incoherent, consistent or inconsistent, and so on, and they possess those properties independently of whether individual scientists or communities of scientists realize it or not. The state of development of scientific theories and the evidence bearing on them at some historical juncture will constitute an objectively existing problem situation. (Chalmers, 1973, p 100.)
Science, then, poses problems. Some scientists, alone or in groups, will appreciate these problems and attempt to tackle them. Other problems will not be discovered. Objective criteria will be constructed to adjudicate between theories, and indeed, Chalmers suggests, objective criteria may exist such that individual scientists and groups can be shown to be mistaken in their assessment of the merits and properties of objectively existing theories.
All four of these views about what constitutes an objective theory: (1) that it is rational and logical (2) that the predictions can be demonstrated to be true (3) that the theory is useful (4) that criteria can be used to adjudicate between theories all imply some form of proof of the theory.
188.8.131.52 The issue of proof
The commonsense notions of proof and fact are invoked to aid the distinction between subjective and objective. Proof supposedly turns subjective knowledge into objective knowledge and turns a subjective statement or view into an objective fact. A theory becomes gains credibility by having been proven.
Proof, as suggested above requires some form of testing, whether it be logical or empirical. Testing out a theory, empirically, in the social sciences is difficult as it is almost impossible to create a controlled environment in which to see, for example, if a theory's predictions are borne out. As a result it is unusual, in the social sciences, to come across suggestions of proven theories. Rather more likely theories are suggested as explanations, for example, but rarely projected as proven. At best theories become 'tried and tested' (which means they seem to work most of the time) or 'assumed' to correspond to reality, often implicitly. Theories that 'work' sometimes become associated with 'common sense'.
One attempt at formulating a version of proof was Weber's (, 1999) requirement that a theory demonstrate 'adequacy' on two levels, 'causal adequacy' and 'meaning adequacy'. Causal adequacy means that the theory is logically consistent with a cause and effect process and that empirically there is some evidence for it. Meaning adequacy refers to a subjective interpretation that is 'typical' of the actions or interpretations that subjects would engage in. Weber ( 1999) states:
A motive is a complex of subjective meaning which seems to the actor himself or to the observer an adequate ground for the conduct in question. We apply the term 'adequacy on the level of meaning' to the subjective interpretation of a coherent course of conduct when and insofar as, according to our habitual modes of thought and feeling, its component parts taken in their mutual relation are recognised to constitute a 'typical' complex of meaning. It is more common to say 'correct.' The interpretation of a sequence of events will on the other hand be called causally adequate insofar as, according to established generalisations from experience, there is a probability that it will always actually occur in the same way. An example of adequacy on the level of meaning in this sense is what is, according to our current norms of calculation or thinking, the correct solution of an arithmetical problem. On the other hand, a causally adequate interpretation of the same phenomenon would concern the statistical probability that, according to verified generalisations from experience, there would be a correct or an erroneous solution of the same problem. This also refers to currently accepted norms but includes taking account of typical errors or of typical confusions. Thus causal explanation depends on being able to determine that there is a probability, which in the rare ideal case can be numerically stated, but is always in some sense calculable, that a given observable event (overt or subjective) will be followed or accompanied by another event.
A correct causal interpretation of a concrete course of action is arrived at when the overt action and the motives have both been correctly apprehended and at the same time their relation has become meaningfully comprehensible. A correct causal interpretation of typical action means that the process which is claimed to be typical is shown to be both adequately grasped on the level of meaning and at the same time the interpretation is to some degree causally adequate. If adequacy in respect to meaning is lacking, then no matter how high the degree of uniformity and how precisely its probability can be numerically determined, it is still an incomprehensible statistical probability, whether dealing with overt or subjective processes. On the other hand, even the most perfect adequacy on the level of meaning has causal significance from a sociological point of view only insofar as there is some kind of proof for the existence of a probability that action in fact normally takes the course which has been held to be meaningful. For this there must be some degree of determinable frequency of approximation to an average or a pure type.
We will explore Weber's approach in more detail in Section 184.108.40.206.1.
More often, these days, theories are not usually considered proven but as providing 'knowledge claims'. A 'knowledge claim' is in essence a plausible and to some extent empirically-supported theory. Richard Rorty has argued that statements or theories designated as 'true' are not accurate descriptions of objective (that is, mind-independent) objects but as useful claims that have managed over time to 'pay their way' (Rorty, 1982), thus providing pragmatic grounds for broad agreement among human investigators.
Thus 'objective knowledge' is, in popular parlance, a well-supported knowledge claim. Bérubé (2005) argues that:
'Objective knowledge' can designate a knowledge-claim having, roughly, the status of being fully supported or proven. Correspondingly, 'subjective knowledge' might designate some unsupported or weakly supported knowledge-claim. It is more accurate to refer to these as objective and subjective judgments, rather than knowledge, but one should be on guard for the use of the term 'knowledge' in this context. This usage fits with the general connotation for the term 'objectivity' of solidity, trustworthiness, accuracy, impartiality, etc. The general connotation for many uses of 'subjectivity' includes unreliability, bias, an incomplete (personal) perspective, etc.
220.127.116.11 Intersubjective agreement: an agreed objectivity?
Does intersubjective agreement prove that there is objective truth? Philosophers do not agree about this proposition. On the one hand, there is the view that intersubjective agreement is as near as one is likely to get to an objective statement. On the other hand, there is a view that even if intersubjective agreement is a proxy for objective knowledge, agreement does not make a statement true or correct.
If two or three or more people agree, for example, that it is cold that does not preclude the possibility of another person claiming that it is not cold. However, the argument goes, that we have much more confidence in our judgments when they are shared by virtually everyone with whom we discuss them than when others disagree. Would we, then, have a high likelihood of objective truth if we had intersubjective agreement among a large number of subjects? Kant, as you may recall from Section 18.104.22.168, was of the view that any claim for objectivity based on intersubjective agreement requires universal agreement. Clearly, though, just because most people agree does not make it correct. Most people agreed the sun went round the earth. More people think god exists than think god is a fiction but that doesn't make god an objective reality and for atheists,who know there is no god, being in a minority doesn't make them any less knowledgeable.
Another line on intersubjective agreement suggests is based on the notion that objective reality is logically consistent. It, therefore, follows that that the atheists' and the believers' knowledge about the existence or otherwise of god cannot both be true; intersubjective disagreement indicates error for at least one of the views. One can also argue that agreement indicates probable truth, because it is unlikely that two people would both be wrong in our judgment regarding an object and both be wrong in exactly the same way. However, this is also a tenuous argument as it is perfectly possible for two or more people to see the same thing incorrectly in the same way: the sun traversing the sky was seen, for centuries, as the sun moving and the earth standing still.
1.7.7 A sterile debate?
Although much time and effort has been spent pondering the subjective or objective nature of social research, the debate is relatively sterile in real world settings. What is needed is not that the research is objective or subjective but that the researcher is able to persuade the audience that the research is valid. In some cases this means claiming the research is 'objective' (as 'subjective' research, which may be valid, nonetheless tends to have negative connotations, implying that it is limited, superficial or biased).
Persuading the audience of the validity of the research may require engaging in rhetoric about data collection methods but, in the last resort, it is the way the data is presented and the way it is linked to other research and integrated into a theoretical analysis that is ultimately persuasive.